When I was a kid I loved to catch funny pronunciations of family and neighbours with my cassetterecorder.
Unfortunately one of the most funniest recording got lost many years ago and from then on it’s a dream of mine to let this recording excist again using the current deep learning technology.
I’ve got two types of audiofiles. The first one is the sample of the voice with a duration of 7 minutes, I think that will be enough to clone a voice.
The second file is the Dutch pronunciations what it’s going to be, including its articulation, but with the voice-sound of the first file.
To be clear, there is no abuse, only for private use.
It is entirely Dutch language.
In total we talk about a file of one minute.
Does anyone have a suggestion where and how I can realise this? is it possible that I send the files so it can be converted to the way I described?
How much will this cost?
Or perhaps I can do it myself with an app like Baidu AI?
I’d like to hear from you.
With kind regards,
Berend from Holland.